Néstor Pérez-Arancibia directs the Micro-Robotic Systems Laboratory at the University of Southern California and places in forgotten mechanical engineering the answer to the challenges of driving and controlling these tiny machines
he fruit of the latest research project by Néstor Pérez-Arancibia , Xiufeng Yang and Longlong Chang, researchers at the University of Southern California, weighs 0.08 grams and is powered by alcohol, specifically methanol. Robeetle – that’s what the creature is called – is a robotic beetle capable of moving without the aid of batteries and carrying charges of up to 0.28 grams, which doesn’t sound like much, but is almost three times its weight. “This is an important step, because it works without cables. No batteries. Autonomously from the energy point of view ”, explains Pérez-Arancibia.
The motivation of this Chilean researcher, who founded the Autonomous Micro-Robotic Systems Laboratory at the University of Southern California in 2013 , is to develop the technologies necessary to create colonies of thousands of insect-scale robots capable of coordinating to carry out useful tasks for the humans. “The many applications for this type of multi-agent systems include artificial pollination, search and surveillance in dangerous environments and data collection in extreme situations,” he says.
These colonies could include crawling insects, such as Robeetle; flyers, like Bee + , another microrobot created in the same laboratory; and even aquatic. But they still have a way to go. “The problem in creating robot insects is energy and power. Right now the best actuators, or the easiest to use, are excited with electricity ”, says the researcher. This means that the inputs that control, for example, the robot’s movements are electrical and involve the use of batteries that, with their limited capacity due to their small size, do not offer enough autonomy.
“There are fundamental restrictions on a physical level. When I’ve spoken to the world’s greatest drum experts, their response is scary and depressing. ” The problem is that it is not clear that it is possible to have batteries with enough energy at such small scales: as it shows, the specific energy of a battery is 1.8 megajoules per kilo, that of the animal fat that living organisms use it amounts to 38 megajoules per kilo. For this reason, Pérez-Arancibia focused on the search for actuators that use sources other than electricity.
Electronic engineering , the undisputed queen of modern technological innovation, falls short of Robeetle . The energy that powers its movements comes from the combustion of methanol, an alcohol that is used as a solvent, antifreeze or, as in this case, a fuel capable of generating 20 megajoules per kilogram. The result is a robot that has more of the steam engine devised by James Watt two centuries ago than a modern gadget . “People forgot about this, but before the electronic age, automatic control was also used. In the steam engine all the controllers were mechanical”, recalls the engineer.
In early engines, the pressure of the water vapor generated the movement and the control of its flows allowed it to be controlled. What happens inside this artificial beetle when it takes a step? In addition to fuel, Robeetle contains an artificial muscle made from a shape memory alloy. This material, subjected to a certain temperature, can remember and adopt a previous form. When methane comes into contact with the platinum-coated surface of the muscle, a chemical reaction occurs that raises the temperature and causes the part to contract.
With this first part we would have managed to generate movement. How do you control to make the successive contractions of the muscle move Robeetle forward? With mechanics. Or, what is the same, with valves that open and close in a coordinated way giving way or stopping the fuel flow. “This is what engineers did before the age of electronics. Control valves without using electronics or computers.”
From the ground to the sky
Robeetle’s limitation is in the rate at which methanol combustion occurs and stops, which sets the pace of the robot’s steps. For a creeping beetle it is enough, but it may not be enough to sustain a bee’s flapping (much faster than a butterfly’s). “We already have ways to do it faster, using different types of fuel,” says the expert. The objective now is to continue increasing that speed, to keep the colony of flying robots that constitutes Pérez Arancibia’s final objective in the air.
Once the mechanical model has been developed and its stability demonstrated, the incorporation of electronic systems that enhance and expand the functions of these micro-robots seems less of a challenge, since Robeetle still has “a lot of space” to put even batteries that power systems. signal processing. “The energy required for that is low. Where most of the energy goes is in the movement. In making the robot walk or fly.”
Robeetle, as proof of concept, has fulfilled its mission to precede that flying robotic colony. But it also has ancestors like the insectocopter designed by the CIA in the 1970s. “They hired Swiss watchmakers to create the mechanisms, but the project failed because the technology was not ready from a manufacturing point of view,” explains the engineer. And it also has cousins in the present, in lines of research closer to biology but also interested in the control of small hybrid organisms. This is the case of grasshoppers modified to detect explosives, developed at the University of Washington, or tuned jellyfish-and consequently able to swim faster- from Stanford University. “It seems very interesting to me, but I couldn’t work on it because two of my doctoral students are vegans,” Pérez-Arancibia says with a laugh.
Will these robots open a new melon when it comes to privacy protection? Will we have to distrust (even more) wasps? “It is no coincidence that, for example, DARPA, the US military projects agency, finances these projects,” admits the researcher. But, before possible military applications, it puts the benefits that it could bring in surgeries, dangerous environments or even when executing hypothetical and futuristic pollination tasks on other planets. “It will be super difficult to have a live bee on Mars,” he reasons. “As with all technologies, there are good things and things that can be sinister associated with them.”